Graph-based prior and forward models for inverse problems on manifolds with boundaries

Graph-based prior and forward models for inverse problems on manifolds with boundaries
复制标题

基于图的先验和前向模型,用于解决带边界流形上的反问题

DOI:
10.1088/1361-6420/ac3994
复制
发表时间:
2022
期刊:
影响因子:
2.1
通讯作者:
Sanz-Alonso, Daniel
Sanz-Alonso, Daniel
中科院分区:
数学2区
文献类型:
--
作者:
Harlim, John;Jiang, Shixiao W;Kim, Hwanwoo;Sanz-Alonso, Daniel

文献摘要

参考文献

被引文献

相似文献

本文发展了带边界流形上偏微分方程约束的贝叶斯反问题数值解的多种学习方法。我们引入了图形化的Matérn型高斯场先验,它允许在边界附近灵活地建模,通过将调和函数与适当的Dirichlet边界条件叠加来表示边值。我们还研究了从PDE参数到观测量的正演模型的基于图的近似。在基于图的先验和正演模型的构造中,我们利用鬼点扩散映射算法来逼近具有经典边界条件的二阶椭圆算子。数值结果验证了我们的基于图的方法,并证明了设计考虑边界条件的先验协方差模型的必要性。
This paper develops manifold learning techniques for the numerical solution of PDE-constrained Bayesian inverse problems on manifolds with boundaries. We introduce graphical Matérn-type Gaussian field priors that enable flexible modeling near the boundaries, representing boundary values by superposition of harmonic functions with appropriate Dirichlet boundary conditions. We also investigate the graph-based approximation of forward models from PDE parameters to observed quantities. In the construction of graph-based prior and forward models, we leverage the ghost point diffusion map algorithm to approximate second-order elliptic operators with classical boundary conditions. Numerical results validate our graph-based approach and demonstrate the need to design prior covariance models that account for boundary conditions.
基于图的贝叶斯半监督学习的一致性和采样算法的可扩展性
DOI: --
发表时间: 2020
影响因子: 6
作者:
Garcia Trillos, Nicolas;Kaplan, Zachary;Samakhoana, Thabo;Sanz-Alonso, Daniel
通讯作者: Sanz-Alonso, Daniel
具有空间白噪声的分数椭圆随机偏微分方程的伽辽金近似的弱收敛性
DOI: 10.1007/s10543-018-0719-8
发表时间: 2017
影响因子: 1.5
作者:
D. Bolin;Kristin Kirchner;M. Kovács
通讯作者: M. Kovács
DOI: 10.1137/19m1295222
发表时间: 2019-10
期刊: ArXiv
影响因子: --
作者:
J. Harlim;D. Sanz-Alonso;Ruiyi Yang
通讯作者: J. Harlim;D. Sanz-Alonso;Ruiyi Yang
DOI: 10.1017/s0962492913000056
发表时间: 2013-01-01
期刊: ACTA NUMERICA
影响因子: 14.2
作者:
Dziuk, Gerhard;Elliott, Charles M.
通讯作者: Elliott, Charles M.
DOI: 10.1137/17m1134214
发表时间: 2018-01-01
影响因子: 2
作者:
Bertozzi, Andrea L.;Luo, Xiyang;Zygalakis, Konstantinos C.
通讯作者: Zygalakis, Konstantinos C.